RAG Playground
Upload a document, watch it get chunked and embedded, then ask questions and inspect exactly which passages grounded the answer.
- LangChain
- Vector DB
- LLM
AI Lab
Client work stays confidential, so the lab exists to show the mechanics: retrieval pipelines, agent graphs, document extraction and AutoML flows you can actually click through. Each build is labelled honestly — live, prototype or experiment.
Upload a document, watch it get chunked and embedded, then ask questions and inspect exactly which passages grounded the answer.
Step through a LangGraph-style agent as it plans, calls tools and recovers from failures — with the state visible at every node.
A sanitised simulation of an outbound sales agent: lead intake, personalised email drafting, reply classification and scheduling, using dummy data.
Paste a video link, get a transcript index and ask context-aware questions with segment references.
Detect, OCR and structure fields from labels and forms — a public-data version of the parcel-processing pipeline.
Drop in a CSV and get profiling, feature selection and a validated regression, forecast or anomaly model — one click.
Render any agent graph as an animated diagram and replay a run node by node.